Mirroring quasi-symmetric organ observations for reducing problem complexity

作者：

Highlights：

• Impact of intra-class variance on machine-learning model complexity is analysed.

• An overview of work performed using quasi-symmetric organ observations is given.

• A method for aligning images of various organs for mirroring orientation is proposed.

• Organ mirroring-alignment accuracy over 99% is achieved on real-world data.

摘要

•Impact of intra-class variance on machine-learning model complexity is analysed.•An overview of work performed using quasi-symmetric organ observations is given.•A method for aligning images of various organs for mirroring orientation is proposed.•Organ mirroring-alignment accuracy over 99% is achieved on real-world data.

论文关键词：Medical image analysis,Within-class variation,Organ orientation,Model complexity,Machine learning

论文评审过程：Received 5 November 2016, Revised 28 April 2017, Accepted 16 May 2017, Available online 18 May 2017, Version of Record 24 May 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.05.041